Ecological analysis of plant community structure and soil effects in subtropical forest ecosystem

Abstract

Background

Subtropical forest plant diversity, characterized by a wide range of species adapted to seasonal variations, is vital for sustaining ecological balance, supporting diverse wildlife, and providing critical ecosystem services such as carbon sequestration and soil stabilization. The Changa Manga Forest, an ecologically rich area with varied vegetation, was analyzed to understand the intricate relationship between plant diversity and environmental factors. This study investigates the diversity patterns, vegetation structure, and environmental influences on forest biodiversity.

Methods

A comprehensive survey was conducted across 127 stands within the Changa Manga Forest to document plant species and classify vegetation communities. Soil samples were collected and analyzed for key physicochemical parameters, while multivariate statistical methods, including hierarchical clustering and ordination, were applied to examine the relationships between vegetation structure and environmental factors. Diversity indices and beta diversity components were calculated to assess variations across plant communities.

Results

The species were classified into six distinct vegetation communities: Neltuma-Ziziphus-Malvestrum (NZM), Broussonetia-Lantana-Morus (BLM), Dalbergia-Lantana-Solanum (DLS), Morus-Abutilon-Ricinus (MAR), Eucalyptus-Vachellia-Sorghum (EVS), and Bombax-Leucaena-Croton (BLC). Analyses using hierarchical clustering and ordination methods revealed significant differences in species composition among these communities, with NZM and DLS exhibiting the highest dissimilarity. Canonical Correspondence Analysis (CCA) indicated that environmental factors such as soil pH, available phosphorus (AP), and organic matter percentage (OM%) are crucial in shaping plant distribution, though the total explained variation remained relatively low. Diversity indices varied significantly among communities, with the NZM community showing the highest Shannon and Simpson diversity, while EVS exhibited the lowest. The beta diversity analysis revealed a high species turnover between certain communities, indicating complex ecological interactions. Our results indicate significant variability in plant community composition and diversity patterns, influenced by edaphic factors and environmental gradients. We anticipate that future environmental changes, such as shifts in soil properties, precipitation patterns, and increased human activity, may exacerbate declines in local plant species richness and disrupt community structures. To preserve the invaluable biodiversity of the study area for future generations, it is essential to implement timely and effective conservation and management strategies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-024-06012-5.

Introduction

Environmental changes are increasingly influencing the patterns and processes of biodiversity assembly, raising concerns about how these shifts might impact the survival [1]. Thus, knowing the abiotic processes that drive species diversity across communities is crucial for monitoring ecosystem stability and structural evolution under climate changethereby creating effective conservation and management measures [2, 3]. The distribution of biomes and forest communities is closely aligned with variations in climate and soil properties [4]. Among these, climate plays a fundamental role in determining the survival, growth, and distribution of plant species, and has consistently been identified as the primary factor explaining diversity patterns in primary forests [5]. Research at continental and global scales underscores the dominant influence of climatic variables on species and functional diversity, species distribution patterns, and the stabilizing effects of plant diversity across spatial scales [6, 7]. Additionally, soil properties, including physical and chemical characteristics such as nutrient availability, pollution levels, salinity, and pH, are crucial for plant growth, survival, biodiversity, and productivity [8]. At the local scale, the influence of soil properties on species distribution can surpass that of climatic variables, especially given the greater spatial variability in soil conditions, which significantly affects local community assembly and turnover [9, 10].

Numerous studies have examined the roles of climate and soil properties in shaping species distribution, diversity, and community assembly [11–13]. However, the interactive effects of these factors on species distribution and diversity within subtropical forest communities remain insufficiently understood [14, 15]. The preferences, responses, and adaptations of plant species to environmental conditions vary widely, leading to complex interspecific relationships and distinct community compositions, which in turn influence local species richness [16]. Furthermore, spatial heterogeneity in environmental factors such as soil nutrients, topography, humidity, temperature, and light conditions likely plays a significant role in determining tree species richness at a local scale in primary forests [17].

Plantations, often established for timber production or land rehabilitation, have shown varying degrees of success in supporting biodiversity, particularly in regions with high environmental variability [18]. Subtropical plantations can serve as critical biodiversity reservoirs, especially in areas where natural habitats have been degraded or lost [19]. Studies in subtropical Asia have highlighted that plantations with diverse structures and species composition can contribute to biodiversity by providing habitats for a variety of native plant species and stabilizing ecological functions [20–22]. Moreover, plantations act as stepping stones for species dispersal and can mitigate the impacts of habitat fragmentation, which is a significant threat in semi-arid regions [23]. The Changa Manga Forest, a managed plantation established in the 19th century, represents one of the largest man-made forests in the world [24]. Although initially created for timber production, this forest now supports diverse plant communities and plays a crucial role in regional biodiversity conservation [25]. Limited studies have explored its contribution to ecological stability and the conservation of native species, underscoring a significant research gap. Recent research in other managed subtropical systems suggests that plantations can support substantial biodiversity, provided they integrate ecological considerations such as species selection, soil management, and spatial heterogeneity [21, 22]. However, the biodiversity potential of the Changa Manga Forest remains poorly understood, particularly concerning its influence on plant community dynamics and their interactions with environmental gradients.

Current research on forest species diversity has largely focused on tropical and temperate forests, while the subtropics, characterized by unique and diverse environmental conditions and forest communities, remain less explored [17, 19]. Subtropical regions, which support extraordinarily high levels of plant diversity and are home to 10–15% of the world’s plant species, host ecosystems of significant ecological importance [22]. These forests are also projected to experience the greatest increase in nitrogen deposition, highlighting their ecological significance and the importance of their service functions [23]. The steep altitudinal gradients, small-scale soil property change, and variable climatic circumstances of subtropical regions make them excellent for studying climatic and edaphic factors affecting tree species diversity [26]. The primary objectives of this study were to (1) elucidate the floristic composition and diversity of plant species within the forest; (2) classify the vegetation into major groups using multivariate classification techniques to delineate distinct plant communities and their relationships to edaphic gradients (3) analyze the influence of edaphic variables on plant distribution patterns, specifically examining how factors effect species composition and community structure. These objectives aim to provide a comprehensive understanding of forest biodiversity and inform conservation stakeholders on the strategies tailored to maintain ecological balance and resilience in the forest ecosystem.

Materials and methods

Study area

The Changa Manga Forest is situated in the subtropical region of central Punjab, Pakistan, within the districts of Kasur and Lahore (Fig. 1). This region falls within a semi-arid, subtropical continental climate zone, characterized by extreme seasonal variations in temperature and humidity [24]. The climate is marked by hot summers, extending from April to September, where temperatures can reach as high as 45 °C, and cool winters from mid-November to the end of January, with temperatures occasionally dropping to near freezing [25]. The monsoon season occurs during July and August, bringing erratic and heavy rainfall, which contributes to the overall annual precipitation of approximately 500–700 mm [24]. The geography of the region is dominated by the vast alluvial plains of the Indus River system, with the Changa Manga Forest itself lying on relatively flat terrain that gently slopes from the northeast to the southwest. The forest is a man-made forest, originally planted in an ecologically dry zone known as ‘Rakh’ during the late 19th century to meet timber demands. Over time, it has evolved into a managed ecosystem characterized by tropical thorn forest vegetation and species adapted to arid conditions. Despite its semi-arid climate, the forest benefits from irrigation provided by the Central Bari Doab Canal system, which supports the growth and regeneration of various tree species, thereby sustaining the forest’s ecological balance and biodiversity.

Fig. 1.

Map showing the Changa Manga Forest in the subtropical region of Punjab Pakistan

Data collection

Field surveys were conducted from January 2021 to December 2023 in the Changa Manga Forest in Punjab, Pakistan. The study aimed to document the vegetation and environmental characteristics of the area. Based on vegetation physiognomy and habitat heterogeneity, 127 sites were randomly selected to represent the majority of vegetation in the forest. At each site, three plots of 100 m² were established randomly, and from each plot 5 quadrats (1 m² for herbs, 5 m² for shrubs, and 10 m² for trees) were placed randomly [4]. Vegetation data were collected by recording the plant species within the quadrat, with measurements taken for species density, cover, and frequency [27–30]. The recorded values for each quadrat were averaged to represent each of the 127 sites.

Subsequently, the Important Value Index (IVI) for each plant species was calculated by determining the relative density, cover, and frequency values [4, 31]. The resulting species matrix, based on the Importance Value Index (IVI) data, was used in all subsequent multivariate statistical analyses. Site richness and diversity were assessed using the averaged density matrix [32]. The vascular plant species recorded during the surveys were identified using the Flora of Pakistan database (http://legacy.tropicos.org/Project/Pakistan), while the taxonomic status of each species was verified through the Plants of the World Online database (https://powo.science.kew.org/), ensuring the use of accepted binomials for improved taxonomic clarity and reliability [33].

Soil analysis

Soil sampling was systematically conducted within each designated plot to facilitate a comprehensive analysis of soil properties. The collection of soil samples for chemical analysis was executed with meticulous precision using polythene bags, which allowed for the extraction of soil mixtures from three distinct points within the 0–20 cm depth range. Soil samples were obtained from three distinct locations within 20 m² quadrats, subsequently combined and thoroughly mixed. A sample of 1 kg of soil was extracted for subsequent physicochemical analysis. Soil texture and pH were determined from soil water extracts using a digital pH meter [34, 35]. The soil organic matter content was quantified according to [36–38]. The quantification of potassium (K) and phosphorus (P) contents was carried out by the standardized procedures elucidated by [39]. A ScalTec moisture analyzer calibrated to 110 °C was used to calculate the moisture content of soil samples. The formula below was used to calculate the saturation percent [40, 41].

Data analysis

Vegetation and environmental data were systematically entered, stored, and processed in Microsoft Excel spreadsheets to meet the requirements of various analytical software programs, including the R statistical package (version 4.3.2; R Core Team 2023) and CANOCO (version 5; [42]). Vegetation groups were classified using agglomerative hierarchical cluster analysis based on species composition similarities, applying Ward’s linkage method with Bray-Curtis distance as the measure. Indicator Species Analysis (ISA) was then employed to identify key indicator species within each predefined vegetation group based on their fidelity values [43]. The vegetation assemblages were subsequently named after the three species with the highest indicator values in each plant association. To assess and compare the β-diversity of communities within the Changa Manga Forest, we conducted a comprehensive analysis using the R package “betapart” [44]. Diversity indices were calculated using PAST software (version 4.12) to assess species richness, evenness, and dominance [45]. The Shannon Index measured diversity by combining richness and evenness, while the Simpson and Dominance indices focused on species dominance. The Evenness Index highlighted the uniformity of species distribution, providing a comprehensive assessment of community diversity. The results were then visually represented using Ridgeline plots, created in OriginPro software, to illustrate the distribution and variation of diversity indices across the study sites. We calculated β-diversity as a ratio of regional to local diversity, focusing on its two primary components: spatial turnover and nestedness-resultant components. Specifically, the spatial turnover was measured using the Simpson pairwise dissimilarity index (βsim), while nestedness was analyzed through the nestedness-fraction of the Sorensen pairwise dissimilarity index (βsne), applying “Sorensen” as the family of dissimilarity index. Detrended Correspondence Analysis (DCA) was used to identify major ecological gradients in species composition, while Canonical Correspondence Analysis (CCA) assessed the influence of environmental variables on species distribution patterns, providing complementary insights into community structure [46, 47]. The significance of vegetation-edaphic relationships was first assessed using the CCA model. When significant differences were found (p < 0.05), individual edaphic variables were further tested. The model was refined through simple and net effects testing of six edaphic variables. To evaluate the strength of associations between plant communities and edaphic properties, we conducted a Mantel test using the Vegan package in R [48]. This test was applied to examine the correlation between species composition and various soil properties across different plant communities.

Results

Floristic composition

During the current study, 168 plant species from Changa Manga Forest Plantation were recorded, and 38 families (Supplementary Table 1). In terms of life span, the forest hosts a significant number of annual species, totaling 105, while biennial species are notably fewer, with only 2 recorded. Perennial species are also prevalent, with 61 identified in the survey. The growth forms of the species in the forest exhibit a diverse array, with herbs being the most abundant at 80 species. Grass species, specifically members of the Poaceae family, account for 18 species, while sedges (Cyperaceae family) and climbers are represented by 6 and 9 species, respectively. Shrubs and sub-shrubs are moderately present with 13 and 6 species each, and tree species form a substantial component of the flora with 36 species identified. Regarding nativity, the forest includes a larger proportion of native species, totaling 113, compared to 55 alien species (Supplementary Table 1).

Fabaceae is the most prominent with 19 genera and 21 species, followed closely by Poaceae with 16 genera and 19 species, and Asteraceae with 13 genera and 15 species. Other notable families include Amaranthaceae and Malvaceae, each hosting 6 and 7 genera respectively, both with 9 species. Solanaceae, Euphorbiaceae, and Moraceae also display significant diversity with 9, 7, and 7 species respectively, spread across 4, 4, and 3 genera. Families such as Convolvulaceae, Cyperaceae, Lamiaceae, and Polygonaceae each comprise 6 species, with varying genera counts. Myrtaceae has 4 species within 3 genera, while families like Apocynaceae, Brassicaceae, Meliaceae, and Verbenaceae each contain 3 species. Less represented families include Bignoniaceae, Cannabaceae, Caryophyllaceae, Menispermaceae, Plantaginaceae, Rhamnaceae, Rubiaceae, and Salicaceae, each having 2 species. Several families are represented by a single species: Acanthaceae, Araceae, Combretaceae, Moringaceae, Nyctaginaceae, Papaveraceae, Petiveriaceae, Phyllanthaceae, Primulaceae, Ranunculaceae, Scrophulariaceae, and Vitaceae, each within a single genus. A strong polynomial relationship (y = 0.0216x² – 1.1968x + 16.919; R² = 0.8811) was observed between the number of species and families, demonstrating a clear trend in species distribution across families (See Fig 2).

Fig. 2.

The relationship observed between species and families in the reported flora. The circle-shaped dots in various colors in the graph show the species distribution pattern in the families. The dotted line in the graph represents the degree of a polynomial function observed in the recorded flora

Major vegetation groups

The major vegetation groups were classified into six groups based on the presence and absence of species through Ward agglomerative cluster analysis (Fig. 3). These groups were named according to the top three indicator species within each group: Neltuma-Ziziphus-Malvestrum community (NZM), Broussonetia-Lantana-Morus community (BLM), Dalbergia-Lantana-Solanum community (DLS), Morus-Abutilon-Ricinus community (MAR), Eucalyptus-Vachellia-Sorghum community (EVS), and Bombax-Leucaena-Croton community (BLC).

Fig. 3.

An agglomerative hierarchical clustering dendrogram of the considered 127 stands representing the six plant species associations in Changa Manga Forest. NZM: Neltuma-Ziziphus-Malvestrum, BLM: Broussonetia-Lantana-Morus, DLS: Dalbergia-Lantana-Solanum, MAR: Morus-Abutilon-Ricinus, EVS: Eucalyptus-Vachellia-Sorghum, and BLC: Bombax-Leucaena-Croton

The Neltuma-Ziziphus-Malvestrum (NZM) community is predominantly represented by Neltuma juliflora (IVI = 42), followed by Ziziphus nummularia (IVI = 21), Malvastrum coromandelianum (IVI = 19.6), Trifolium alexandrinum (IVI = 10.5), and Medicago polymorpha (IVI = 6.4). Other notable species within this community include Chenopodium album (IVI = 4.7), Janochloa antidotalis (IVI = 4.5), Oxalis pes-caprae (IVI = 3.3), Eclipta prostrata (IVI = 3.3), Erigeron canadensis (IVI = 3.2), Cannabis sativa (IVI = 3), and Achyranthes aspera (IVI = 2). This plant community comprised of 15 samples.

The Broussonetia-Lantana-Morus (BLM) community is dominated by Broussonetia papyrifera (IVI = 57.5), Lantana camara (IVI = 41), and Morus alba (IVI = 10). Additionally, Sida rhombifolia (IVI = 4.664), Abutilon indicum (IVI = 4.6), Parthenium hysterophorus (IVI = 4), Lathyrus aphaca (IVI = 3.9), Solanum virginianum (IVI = 3.41), and Polygonum aviculare (IVI = 3) contribute to the biodiversity of this community. Other significant species include Eragrostis pilosa (IVI = 2.5), Rumex obtusifolius (IVI = 2.6), and Vitex negundo (IVI = 2). This plant community comprised of 16 samples.

The Dalbergia-Lantana-Solanum (DLS) community features Dalbergia sissoo (68.7) as the dominant species, followed by Lantana camara (IVI = 37.8) and Solanum viarum (IVI = 15.2). Other important species include Moringa oleifera (IVI = 2.6), Rumex obtusifolius (IVI = 4.053), Solanum virginianum (IVI = 3.9), Lathyrus aphaca (IVI = 3.8), and Lepidium didymum (3.8). Additionally, Abutilon indicum (3.5), Hamelia patens (3), Abutilon theophrasti (IVI = 2.9), and Abutilon hirtum (IVI = 2.2) are significant contributors to the DLS community ecological framework, illustrating a rich diversity of flora. This plant community comprised of 17 samples.

The Morus-Abutilon-Ricinus (MAR) community is primarily composed of Morus alba (IVI = 42.6), Abutilon indicum (IVI = 28.3), and Ricinus communis (IVI = 25). The presence of species such as Calotropis procera (IVI = 6.8), Sida rhombifolia (IVI = 6.7), Psidium guajava (IVI = 6.1), and Salvia plebeia (IVI = 5.315) further enriches this community. Other noteworthy species include Solanum erianthum (IVI = 3.7), Solanum incanum (IVI = 3.7), Lantana camara (IVI = 3.2), Solanum viarum (IVI = 3), and Physalis peruviana (IVI = 2.8), contributing to the MAR community’s distinctive ecological identity. This plant community comprised of 18 samples.

In the Eucalyptus-Vachellia-Sorghum (EVS) community, Eucalyptus camaldulensis (IVI = 36.3) is the predominant species, followed by Vachellia farnesiana (IVI = 23.6) and Sorghum halepense (IVI = 14.9). Significant contributions from species such as Achyranthes aspera (IVI = 11.9), Neltuma juliflora (IVI = 10.2), Medicago polymorpha (IVI = 8), and Polygonum plebeium (IVI = 6.6) define this community. Additionally, Cannabis sativa (IVI = 4.78), Euphorbia heterophylla (IVI = 4.5), Panicum turgidum (IVI = 4.4), Eragrostis barrelieri (IVI = 4.1), and Rumex spinosus (IVI = 4) further contribute to the EVS community ecological composition. This plant community comprised of 20 samples.

The Bombax-Leucaena-Croton (BLC) community features Bombax ceiba (IVI = 46.5) as the dominant species, with Leucaena leucocephala (IVI = 38.9) and Croton bonplandianus (IVI = 17.5) also playing significant roles. Other key species include Medicago polymorpha (IVI = 14.8), Silybum marianum (IVI = 10.9), Malvastrum coromandelianum (IVI = 10.4), Janochloa antidotalis (IVI = 10.4), and Populus alba (IVI = 9.8). The presence of Syzygium cumini (IVI = 9.4), Lactuca serriola (IVI = 9.3), Xanthium strumarium (IVI = 8.2), and Vachellia nilotica (IVI = 7.9) further underscores the diversity and ecological interactions within the BLC community. This plant community comprised of 26 samples.

Diversity pattern

The Tukey post hoc test for the Shannon Index reveals notable differences among the communities. The NZM community exhibits a significantly higher mean Shannon diversity (1.58 ± 0.06) compared to BLM and MAR. DMS displays the highest mean Shannon diversity (1.92 ± 0.10), significantly greater than NZM and BLM. BLC also shows a significantly higher Shannon diversity (1.73 ± 0.06) compared to MAR, EVS, and BLM. The comparisons indicate that EVS has the lowest Shannon diversity (1.06 ± 0.03), significantly lower than NZM, DMS, and BLC. For the Simpson Index, significant differences are observed among several communities. The NZM community has a significantly higher Simpson diversity (0.85 ± 0.02) than BLM, MAR, EVS, and BLC. The DMS community exhibits a higher Simpson diversity (0.87 ± 0.02) compared to BLM and MAR. BLC (0.82 ± 0.01) shows significantly higher Simpson diversity than MAR and EVS but is not significantly different from NZM and DMS. EVS has the lowest Simpson diversity (0.50 ± 0.03), significantly lower than all other communities except BLM (Fig. 4).

Fig. 4.

Variation of diversity indices among the six plant communities in Changa Manga Forest. Figures represent ridgeline plots with raw data. Lowercase letters on the left indicate statistically significant differences determined by estimated marginal means. The Y-axis is displayed according to community classifications. NZM: Neltuma-Ziziphus-Malvestrum, BLM: Broussonetia-Lantana-Morus, DLS: Dalbergia-Lantana-Solanum, MAR: Morus-Abutilon-Ricinus, EVS: Eucalyptus-Vachellia-Sorghum, and BLC: Bombax-Leucaena-Croton

The Dominance Index indicates significant variations between communities. NZM shows a significantly lower dominance (0.21 ± 0.01) compared to BLM, DMS, and EVS. BLM has a significantly higher dominance (0.60 ± 0.05) compared to NZM and DMS, but is not significantly different from MAR and BLC. EVS (0.39 ± 0.05) has significantly higher dominance than MAR but is not significantly different from DMS and BLC. BLC shows no significant difference in dominance compared to NZM, DMS, and MAR. The Evenness Index results reveal that MAR has a significantly higher evenness (1.31 ± 0.09) compared to NZM, BLM, DMS, and EVS. BLM also shows significantly higher evenness (1.17 ± 0.09) compared to NZM and EVS. DMS (0.90 ± 0.07) exhibits significantly lower evenness compared to BLM and MAR. BLC (0.92 ± 0.02) shows significantly lower evenness compared to MAR and BLM but is not significantly different from NZM and DMS. EVS (0.93 ± 0.01) does not show significant differences in evenness compared to NZM and DMS but is significantly lower than MAR. The Tukey post hoc test highlights significant variations in diversity indices across the different communities, with DMS often showing the highest diversity and EVS the lowest across various indices (Fig. 4).

Beta diversity patterns

The beta diversity analysis of plant communities within Changa Manga Forest reveals significant variability in species composition among the six identified communities. The Simpson dissimilarity index (βsim) indicates the highest dissimilarity between the NZM community and both the DLS and MAR communities, each scoring a value of 1. This suggests a complete turnover of species between these communities. The EVS community also shows complete dissimilarity with DLS and MAR, each scoring 1. In contrast, the BLM community exhibits moderate dissimilarity with the NZM community at 0.9 and lower dissimilarity with the EVS community at 0.8. The BLC community shows the least dissimilarity with the NZM community at 0.4 (Fig. 5). The nestedness-resultant dissimilarity index (βsne) presents low values overall, indicating minimal nestedness among the communities. The highest value of 0.3 is observed between the MAR and BLC communities, suggesting some degree of nestedness but still relatively low compared to overall beta diversity. The values between NZM and the other communities are particularly low, ranging from 0 to 0.2, indicating that the unique species turnover is more influential in driving beta diversity rather than nestedness.

Fig. 5.

Dissimilarity clusters based on spatial turnover (βsim) and nestedness-resultant components (βsne) of beta diversity among the six plant communities in Changa Manga Forest (NZM: Neltuma-Ziziphus-Malvestrum, BLM: Broussonetia-Lantana-Morus, DLS: Dalbergia-Lantana-Solanum, MAR: Morus-Abutilon-Ricinus, EVS: Eucalyptus-Vachellia-Sorghum, and BLC: Bombax-Leucaena-Croton)

Ordination of communities

The results of the Detrended Correspondence Analysis (DCA) revealed a significant distribution of variation across the first four axes. The eigenvalues for Axis 1, Axis 2, Axis 3, and Axis 4 were 0.976, 0.884, 0.731, and 0.610, respectively, indicating a gradual decrease in explained variance. The total inertia, representing the overall variance in the dataset, was 30.261. The lengths of the ecological gradient captured by each axis were 11.772 for Axis 1, 9.206 for Axis 2, 7.959 for Axis 3, and 6.984 for Axis 4, suggesting that the first axis captured the most substantial variation in species composition (Table 1). The cumulative percentage variance explained by the species data across the axes increased progressively, with Axis 1 explaining 45.3%, Axis 2 explaining 32.1%, Axis 3 explaining 15.0%, and Axis 4 explaining 10.2% (Fig. 6). This indicates that while the first two axes capture a significant portion of the variation, the overall explained variance remains distributed across multiple dimensions, suggesting complex underlying ecological gradients.

Table 1.

Summary of Detrended Correspondence Analysis (DCA) results showing eigenvalues, lengths of ecological gradients, and cumulative percentage variance explained for the first four axes

Axis	Axis 1	Axis 2	Axis 3	Axis 4
Eigenvalues	0.976	0.884	0.731	0.61
Lengths of Gradient	11.772	9.206	7.959	6.984
Cumulative % Variance	45.3	32.1	15	10.2

Fig. 6.

Results of the Detrended correspondence analysis (DCA) showing the distribution of species and six plant communities (NZM: Neltuma-Ziziphus-Malvestrum, BLM: Broussonetia-Lantana-Morus, DLS: Dalbergia-Lantana-Solanum, MAR: Morus-Abutilon-Ricinus, EVS: Eucalyptus-Vachellia-Sorghum, and BLC: Bombax-Leucaena-Croton)

Impact of edaphic variables on plant distribution

The Canonical Correspondence Analysis (CCA) results reveal that the total variation in the dataset is 29.5, with the explanatory variables accounting for 6.1% of this variation (Fig. 7). After adjustment, the explained variation is reduced to 1.5%. The analysis shows that the first axis captures the highest eigenvalue (0.4854) and explains 1.64% of the variation, with a cumulative explained variation of 4.67% across all four axes. The pseudo-canonical correlations for the axes are relatively high, ranging from 0.7606 to 0.8283, indicating a strong relationship between the species data and the environmental variables. In the simple term effects analysis, all variables except soil texture, EC, and saturation percentage have statistically significant effects on species distribution (p < 0.05). AP (1.5%, p = 0.002), OM% (1.4%, p = 0.002), pH (1.3%, p = 0.002), and AK (1.1%, p = 0.002) are the most influential factors (Table 2). Although soil texture, EC, and saturation percentage explain about 1% of the variation, their effects are not consistently significant. The conditional term effects analysis further refines these findings. AP remains the most significant variable (1.5%, p = 0.002), followed by pH (1%, p = 0.006), and soil texture (1.1%, p = 0.012). Saturation percentage, AK, and OM% have weaker and non-significant contributions in this context, indicating that while they contribute to species-environment relationships, their independent effects are minimal when controlling for other variables. The CCA results highlight the critical role of AP, OM%, pH, and AK in shaping species distributions, with soil texture and EC playing more minor roles. The analysis underscores the complexity of species-environment interactions, where certain variables have prominent effects while others contribute less significantly to the overall explained variation.

Fig. 7.

Canonical Correspondence Analysis (CCA) results show the relationship between edaphic variables and species distributions

Table 2.

Summary of simple and conditional term effects from Canonical Correspondence Analysis (CCA). The table shows the percentage of variation explained by each environmental variable, along with the corresponding pseudo-F values and p-values. Simple-term effects indicate the individual influence of each variable, while conditional-term effects represent the influence of each variable after accounting for the effects of others. Significant p-values (p < 0.05) suggest a meaningful contribution of the environmental variable to species variation

Analysis	Name	Explains %	pseudo-F	P
Simple Term Effects:	AP	1.5	1.9	0.002
	OM%	1.4	1.8	0.002
	pH	1.3	1.7	0.002
	AK	1.1	1.4	0.002
	Soil texture	1	1.3	0.084
	EC	1	1.3	0.026
	Saturation %	1	1.2	0.076
Conditional Term Effects:	AP	1.5	1.9	0.002
	pH	1	1.3	0.006
	Soil texture	1.1	1.3	0.012
	Saturation %	0.9	1.1	0.132
	AK	0.9	1.1	0.12
	OM%	0.8	1	0.438

Correlation between environmental variables and species groups

The Mantel test results revealed varying degrees of correlation between environmental variables and species groups. For the NZM community, most environmental factors, such as pH (r = 0.055, p ≥ 0.05) and EC (r = 0.041, p ≥ 0.05), showed weak correlations with insignificant p-values (Fig. 8). However, some factors like OM% (r = 0.066, p < 0.01), AP (r = 0.087, p < 0.01), and AK (r = 0.056, p < 0.01) displayed significant yet weak positive correlations. The BLM community exhibited consistently weak correlations across all environmental variables, with no significant results (all p ≥ 0.05). Similarly, the DMS and MAR communities showed weak and non-significant correlations for all tested ecological factors. While showing weak correlations, the EVS community did not yield any significant p-values either. For the BLC community, only soil saturation (%) showed a weak but statistically significant positive correlation (r = 0.149, p = 0.049), whereas all other variables presented weak and non-significant correlations (p ≥ 0.05) (Fig. 8). Overall, the Mantel test indicates that while there are some significant correlations between certain environmental factors and species groups, these relationships are generally weak, suggesting that other factors might be influencing species distribution patterns.

Fig. 8.

Correlation matrix between environmental variables and species groups, overlaid with Mantel test results. The plot displays Pearson’s correlation coefficients as color-coded squares, with significant Mantel test correlations highlighted by curved lines. Line thickness represents the strength of Mantel’s r values, while color indicates the significance of Mantel’s p values. NZM: Neltuma-Ziziphus-Malvestrum, BLM: Broussonetia-Lantana-Morus, DLS: Dalbergia-Lantana-Solanum, MAR: Morus-Abutilon-Ricinus, EVS: Eucalyptus-Vachellia-Sorghum, and BLC: Bombax-Leucaena-Croton

Discussion

The current study identified 168 plant species across 38 families, highlighting its rich biodiversity and ecological significance for conservation and ecological studies. The high diversity aligns with findings in other tropical forests, where a broad range of species reflects the complex interactions within these ecosystems [4, 49, 50]. The prevalence of annual species within the forest suggests a dynamic ecosystem, with species constantly adapting to changing environmental conditions, a phenomenon also observed in other studies of tropical ecosystems [51, 52]. The low number of biennial species may indicate that the forest environment does not favor species with two-year life cycles, possibly due to the competitive dominance of perennials that thrive in stable conditions [53]. The forest structural complexity is evident in the dominance of trees and herbs, which support a variety of ecological niches [54]. This mirrors global patterns where diverse forest strata contribute to high biodiversity [55]. The substantial presence of herbs is consistent with previous studies, which highlighted the dominance of herbaceous species in many subtropical managed forests [56, 57]. The presence of climbers and grasses suggests varied habitat conditions within the forest, accommodating both open and restricted environments [58]. In terms of species nativity, the forest harbors more native species than alien species, suggesting that non-native species have not yet significantly altered the forest ecological integrity. However, the presence of alien species highlights the ongoing threat of invasive species, which can disrupt native communities by outcompeting indigenous species for resources [59]. The floristic composition of the forest was dominated by the Fabaceae, which aligns with global trends where Fabaceae nitrogen-fixing ability is crucial for ecosystem functions [60]). The prominence of Poaceae and Asteraceae reflects the typical mix of grasslands and herbaceous ecosystems in tropical forests [61]. The resilience of Amaranthaceae and Malvaceae suggests the forest’s adaptability to climate change [62], while the diversity within Solanaceae, Euphorbiaceae, and Moraceae further highlights its ecological richness [63].

The classification of major vegetation groups in the Changa Manga Forest into six distinct communities provides a comprehensive understanding of the forest ecological structure. The NZM community, characterized by species such as Neltuma juliflora and Ziziphus nummularia, aligns with findings from similar dry forest ecosystems, where these species are often dominant due to their adaptability to arid conditions [64]. The significant presence of Malvastrum coromandelianum and other herbaceous species within this community suggests a well-developed understory, consistent with studies highlighting the role of herbs in maintaining ecosystem dynamics [65]. The BLM community, dominated by Broussonetia papyrifera and Lantana camara, illustrates the influence of invasive species in altering native plant communities, a phenomenon widely documented in tropical and subtropical forests [66]. These species’ presence indicates anthropogenic disturbances, which have facilitated their spread and dominance over native flora [67]. The DLS community, with Dalbergia sissoo as the predominant species, reflects the patterns observed in riverine and moist deciduous forests [68]. Dalbergia species play a critical role in maintaining soil fertility and supporting diverse plant communities [69]. However, the coexistence of Lantana camara in this community emphasizes the ongoing challenge of managing invasive species in forest conservation efforts [70]. The MAR community, where Morus alba and Ricinus communis are key species, represents a transitional zone between cultivated lands and forested areas. This community composition is typical of disturbed habitats, where generalist species thrive, often at the expense of more sensitive native species [71]. The EVS community, dominated by Eucalyptus camaldulensis, highlights the impact of monoculture plantations on forest biodiversity [72]. The lower diversity indices observed in this community are consistent with studies showing that Eucalyptus plantations often lead to reduced understory diversity and altered soil conditions, which can negatively affect native species [73]. The BLC community, with Bombax ceiba as a dominant species, reflects the typical structure of semi-evergreen forests, where large canopy trees create a microhabitat conducive to a diverse assemblage of understory species [74]. The presence of Leucaena leucocephala, however, suggests that even these relatively stable communities are not immune to the effects of invasive species [75].

The diversity indices further reinforce the differences between these communities. The higher Shannon and Simpson diversity observed in the DLS and LC communities suggest a complex plant assemblage; however, this may also be influenced by the presence of invasive species, whereas the lower diversity in the EVS community reflects the homogenizing effect of Eucalyptus plantations. These findings are consistent with global studies that highlight the negative impact of monocultures on biodiversity [76, 77]. The dominance and evenness indices provide additional insights, with higher dominance in communities like BLM and EVS suggesting a lack of species balance, often a sign of ecological stress or disturbance [78]. The vegetation groups identified in the forest offer a snapshot of the forest ecological diversity and the ongoing challenges it faces [79]. The presence of invasive species, the effects of monoculture plantations, and the varying levels of species diversity across communities underscore the importance of targeted conservation efforts to preserve the ecological integrity of this unique forest [4, 80, 81]. These findings contribute to a broader understanding of how different environmental and anthropogenic factors shape forest ecosystems, providing valuable insights for future conservation and management strategies [82]. The results show notable variations in the Shannon and Simpson diversity indices across communities, reflecting differences in species diversity and evenness. These findings align with recent research demonstrating that communities can exhibit distinct diversity profiles [83–85]. The findings demonstrate the need of utilizing both the Shannon and Simpson indices to gauge the evenness and variety of a community. There are notable differences in the distribution of species and community structure between the various communities, as indicated by the Dominance Index and Evenness Index values. In line with recent studies emphasizing the value of maintaining varied and even communities, the low dominance and high evenness seen in NZM and MAR communities point to a fairer distribution of species [86]. On the other hand, the BLM and EVS communities exhibit poor evenness and high dominance, suggesting a more unequal individuals within species. This might potentially be a sign of environmental stresses or disturbance regimes [87]. The diversity indices highlight that species evenness and richness are well-distributed within the studied communities, indicating a balanced ecological structure without dominance by a few species [88]. According to recent studies, several diversity indices should be considered in order to capture various elements of biodiversity [89]. The results demonstrate the need to utilize both the evenness and dominance indices to analyze species distribution and community structure.

According to beta diversity analysis, significant variation in species composition is seen among the six identified plant communities in the forest. The Simpson dissimilarity index (βsim) illustrates the total species turnover that occurs between some communities, such as DLS/MAR and EVS, and NZM and DLS/MAR. This implies that there is little overlap between the species compositions of both communities [90]. By comparison, some community pairs such as BLM and NZM/EVS and BLC and NZM show moderate to low dissimilarity. This suggests that although these ecosystems have distinct compositions, they share some species [91]. The lowest nestedness is seen among communities according to the nestedness-resultant dissimilarity index (βsne), with MAR and BLC having the greatest value (0.3). This implies that species composition is a distinct mix of species rather than merely a subset of another community [92]. The low βsne values indicate that unique species turnover, rather than nestedness, promotes beta diversity, especially between NZM and other communities. This is in line with previous research that demonstrates a variety of variables, including species interactions, environmental gradients, and disturbance regimes, which can affect beta diversity [90, 91]. According to recent research [85], comprehending the intricate patterns of species turnover and nestedness in plant communities requires considering a variety of beta diversity components.

The results of the CCA and Mantel tests offer a detailed yet complex picture of the impact of edaphic variables on plant distribution. The overall explained variation in the dataset by these environmental variables is quite low, indicating that while they play a role in shaping species distributions, their influence is limited when considered in isolation. This finding is consistent with the broader ecological literature, which emphasizes the multifactorial nature of species distribution patterns, where numerous abiotic and biotic factors interact to determine the presence and abundance of plant species [93, 94]. This aligns with previous studies, as mentioned by [43] and [95], which argue that while soil properties are critical, they often interact with other environmental gradients, such as climate and topography, to influence species distributions. The significant influence of available phosphorus, organic matter percentage (OM%), pH, and available potassium on species distribution is consistent with the established understanding of plant ecology. Phosphorus availability, for instance, is well-known to be a limiting factor in many terrestrial ecosystems, particularly in nutrient-poor soils, where it can significantly affect plant growth and community composition [95, 96]. The importance of soil pH in influencing nutrient availability and microbial activity has been extensively documented [97, 98], and the role of organic matter in enhancing soil fertility and structure is similarly well-understood [99]. The findings show that AP, OM%, and pH are among the most influential variables in this study. This corroborates with the results of previous research in various ecosystems, including tropical forests [100–103].

However, the relatively weak influence of soil texture and EC contrasts with some earlier studies that have highlighted the importance of these factors in specific contexts [104, 105]. For example, soil texture has been shown to affect water retention and drainage, which are critical for plant survival in arid and semi-arid regions [106]. The limited impact of EC in this study might reflect the specific characteristics of the study area, where salinity is not a major limiting factor for most plant species [25]. This observation suggests that the importance of edaphic variables may vary significantly across different ecosystems and that their effects may be context-dependent, influenced by local environmental conditions and species adaptations [107]. The Mantel test results further complicate the picture, revealing strong correlations between edaphic variables and species groups across most of the communities studied [108]. This finding indicates that the selected edaphic variables may not be the primary drivers of species distribution in these communities. The weak yet significant correlations observed for certain variables, such as OM%, AP, and AK in the NZM community, suggest that while these factors contribute to species distribution, their influence is relatively minor compared to other potential factors, such as interspecific competition or microhabitat heterogeneity [109, 110]. The results of this study underscore the complexity of species distribution patterns and the limitations of focusing solely on edaphic variables. While factors such as AP, OM%, pH, and AK have clear influences on plant distribution, and their effects are relatively modest when considered in isolation.

Recommendations

To effectively manage and conserve forest biodiversity in subtropical regions, it is essential to implement a comprehensive approach that addresses both ecological and anthropogenic factors influencing these ecosystems. The study findings highlight the critical role of soil health, particularly variables such as available phosphorus, organic matter (OM%), and pH, in shaping plant species distributions. Therefore, forest management practices should prioritize soil conservation strategies, including regular soil health assessments and the enhancement of soil fertility. Conservation efforts should focus on mitigating the impacts of land use changes, deforestation, and other human activities that threaten biodiversity. Integrating community involvement in conservation initiatives can foster sustainable land use practices and enhance local stewardship of forest resources. Furthermore, adaptive management strategies should be developed to account for the potential impacts of climate change, such as shifts in temperature and precipitation patterns, which could further influence species distribution and ecosystem stability. Protecting key habitats and promoting reforestation in degraded areas are also vital for maintaining biodiversity. By addressing these factors, conservation and management efforts can more effectively safeguard the unique biodiversity of subtropical forests, ensuring their resilience and ecological function for future generations. Future research should focus on long-term monitoring of environmental changes and their impacts on plant communities to develop adaptive management practices that safeguard the forest’s ecological integrity.

Conclusions

This study provides a comprehensive analysis of the plant community structure and edaphic influences within the subtropical forest ecosystem of Changa Manga Forest Plantation. Our findings reveal a rich and diverse flora, with a total of 168 plant species distributed across 38 families, highlighting the forest’s significant botanical diversity. The application of multiple diversity indices, including Shannon, Simpson, and Evenness indices, underscores considerable variation in community diversity and dominance across these plant assemblages. Furthermore, our analysis of beta diversity and ordination results indicates substantial turnover and distinct species composition among the communities, reflecting the complex ecological gradients present in the forest. The CCA highlights that key edaphic variable such as available phosphorus, organic matter percentage, and pH significantly influence plant distribution, with other variables like soil texture and electrical conductivity having lesser effects. These insights suggest that while soil characteristics play a notable role in shaping plant communities, additional factors may also contribute to observed patterns. In light of these results, it is evident that the subtropical forest’s plant diversity is intricately linked to its environmental conditions. To preserve this valuable biodiversity and ensure the continued health of the forest ecosystem, targeted conservation and management strategies are essential.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization, M.W. and S.S.; Methodology, S.S; software, M.W. and S.S.; validation, F.A. and S.I.; formal analysis, M.W.,; investigation, M.W.; resources, F.A. and S.F.; Data curation, A.A., B.F.A.,; writing—original draft preparation, M.W., S.F; writing—review and editing, M.W., F.A., A.A., R.S., S.F, and B.F.A.,; visualization, M.W.; supervision, S.S. and M.W.; project administration, S.S.; funding acquisition A.A. All authors having substantial contributions in research, read and agreed to the published version of the manuscript.

Funding

The authors extend their appreciation to the Deanship of Research and Graduate Studies at King Khalid University for funding this work through Large Research Project under grant number RGP2/56/45.

Data availability

Data is contained within the article.

Declarations

Ethical approval

Not applicable.

Consent for publication

Not applicable—this manuscript has no personal data from the authors.

Clinical trial number

Not applicable.

Competing interests

The authors declare no competing interests.